Retrieval-Augmented In-Context Learning for Domain Adaptation

DAICL: a retrieval-augmented in-context learning framework for unsupervised domain adaptation that retrieves target examples and jointly optimizes task and LM.

Voice Chips: Bluetooth Features and Use Cases

Overview of voice recognition chips and Bluetooth features, comparing WTK6900HA8 and WTK6900HC with power, BLE/audio modes and application guidance.

PanopticNeRF-360: Panoptic Labels from 3D Annotations

PanopticNeRF-360 uses coarse 3D annotations and NeRF-based joint geometry-semantic optimization to synthesize RGB images and panoptic segmentation from novel views.

Gitee Repository Assistant with Java and MCP

Integrating Gitee MCP with Java using LangChain4j to build a Spring Boot AI repository assistant, demonstrating stdio and SSE transport modes.

Probability Distributions for Machine Learning

Technical overview of probability distributions for urban research, explaining random variables and density functions (PMF, PDF, CDF) and common discrete cases like Bernoulli.

DeepSeek and Four Key Factors for Large Model Applications

Examines the 2025 surge in large models after DeepSeek R1, detailing application value beyond chat and four key elements: domain expertise, knowledge base, agents, and engineering.

Running the KleidiAI Microkernel on Bare Metal

Run KleidiAI matmul microkernel on bare metal and compare Arm toolchains (AC6, GCC, ATfE) with cycle-count benchmarks and compiler optimization analysis.

Smart Video Analytics with Edge Computing and Industrial PCs

Overview of intelligent video analytics (IVA), deployment scenarios and why edge computing and industrial computers enable low-latency, secure real-time video AI.

Implicit Harmful Outputs in Large Language Models

Tsinghua study analyzes implicit harmful content in LLMs, showing how SFT and RLHF can induce subtle abusive outputs that evade harmful-content detectors.

AI's Impact on Industrial Software

Siemens Digital Industries white paper on AI in electronic systems design, exploring AI methods and applications for PCB design, component selection, layout and verification.

How Far Are VLMs from Visual Deductive Reasoning?

Evaluation of VLMs on visual deductive reasoning using Raven's Progressive Matrices (RPMs); analyzes perception bottlenecks, prompt effects, and error modes.

How Computer Systems Address Challenges of Large AI Models

Dr. Mao Yang examines how computer systems must evolve - ultra-large-scale computing, cloud redesign, and distributed systems - to support large models and next-gen AI

Do Larger Vision Models Always Perform Better?

Scale scaling (S2): run pretrained frozen vision models at multiple image resolutions to produce multi-scale representations, matching larger models on visual tasks.

IMX8M Plus-Based Smart Pharmacy Vending Machine Design

Technical overview of smart pharmacy vending machines using an i.MX8M Plus carrier board: hardware, 2.3 TOPS NPU, connectivity, I/O, displays, storage, and OS support.

Deep Learning Beginner Demo: Handwritten Digit Case

Overview of deep learning fundamentals and a TensorFlow 2 handwritten digit recognition demo, covering neurons, activation/loss functions, CNNs, training, and prediction.

Efficient Deployment of DeepSeek on Intel Platforms

Guide to deploying the DeepSeek LLM on Intel Arc platforms: hardware, BIOS, Ubuntu 24.10, Arc B580 drivers and OpenVINO setup to validate model inference.

LLM Programming: Lessons from Redis's Creator

Programmer's account of using large language models to accelerate Python one-off scripts and ONNX model interpretation, and limits in systems programming.

High-Flexibility, High-Precision Bionic Robotic Hand

Technical overview of a modular, safety-focused bionic robotic hand integrating ROS, cable-driven mechanics, multi-modal sensing and AI-based control for adaptive manipulation.

RDK X3 Campus History Guide Robot with SLAM Navigation

Technical overview of an interactive guide robot for campus history museums: SLAM-based navigation, multi?mic voice interaction, modular chassis and AI-driven commentary.

TDK spin photodetector advances optical detection

TDK's spin photodetector using MTJ magnetic elements enables wide-spectrum, ultra-fast optical detection and flexible PIC integration for high-speed links.